The Real Odds

Three large studies get quoted constantly about memecoin odds, usually one at a time, usually to prove whatever the writer already believed. Put side by side, they do not contradict each other - they measure different things.

A Dune dashboard covering 13.4 million pump.fun wallets (reported by Decrypt, January 2025) found that 55,012 wallets, 0.4% of the total, had realized more than $10,000 in profit, and about 294 had realized more than $1 million. It cannot see trades after graduation, coins still held, or the roughly 30% of wallets that were bots that sold once.

CoinGecko Research, covering 3.1 million wallets active in April 2026, found 73.28% in realized profit: 65.14% had made $1 to $500, and 5.37% had made more than $1,000. It explicitly cannot see bagholders who never sold a token even as it fell to zero, and it does not filter out bots or wash trading.

Solidus Labs, covering over 7 million pump.fun tokens from January 2024 to March 2025, found that 98.6% fell below $1,000 of liquidity. It cannot tell you whether any individual collapse was planned.

Read together: nearly every coin dies, and most wallets that actually sell come out a few dollars ahead or behind, because most people buy small and sell fast - a loss only counts once you sell, and the dead coins still sitting in wallets are invisible to both profit studies. A small number of wallets make very large sums, and none of these studies can tell you how many of those are bots or the coins' own creators.

If your plan is to copy a successful wallet, one more study matters. A paper accepted at the 2026 Web Conference proves, from the bonding curve's own formula, that a copier structurally overpays on every replicated buy, because the copied wallet's purchase has already moved the price. In the authors' tests on 6,000 coins, even correctly identifying profitable wallets to copy produced negative returns for the copier; their own, more elaborate selection method averaged 14% for the original wallet and an estimated 3% for the copier - the best case in the paper, before anyone reacts to being copied.

On being paid to post about it: well-known traders are paid in more ways than trading. Axiom's documentation states it pays a referrer 30% of its net fee on every trade made by everyone they refer, plus 3% and 2% on the two levels below that - paid whether the referred trader wins or loses. And the leaderboard most people use to rank these traders, Kolscan, was bought by pump.fun itself in July 2025. A leaderboard ranks wallets someone has labelled; the house owns this one.